Behavior part 1: Transportation

Questions 7, 8, 9, 10. Transportation behavior.

## Create survey object.
options(digits = 4)
options(survey.lonely.psu = "adjust")

des <- svydesign(ids = ~1, weights = ~weight, data = df[is.na(df$weight)==F, ])

Q7. What types of public transportation do you regularly use?

# subset question data, rename columns, gather into single column
q7_df <- df %>%
  select(CaseID, PPGENDER, PPAGE, ppagecat, PPETHM, PPINCIMP, PPEDUC, PPEDUCAT,
         work, PPWORK, marital, PPMARIT, PPMSACAT, ppreg9, PPSTATEN, PPHOUSE, PPRENT, PPNET,
         Q7_1:Q7_7, Q7_otherText, weight) %>%
  rename(Bus = Q7_1, Carpool = Q7_2, Subway = Q7_3, Train = Q7_4,
         Taxi = Q7_5, Airplane = Q7_6, Other = Q7_7) %>%
  gather(Q7_q, Q7_r, Bus:Other, na.rm = T) %>%
  mutate(Q7_q = as.factor(Q7_q))

# select only Yes responses
q7_df <- q7_df[(q7_df$Q7_r)=='Yes', ]
str(q7_df)

'data.frame':   324 obs. of  22 variables:
 $ CaseID      : int  2 5 15 18 23 44 53 69 72 75 ...
 $ PPGENDER    : Factor w/ 2 levels "Female","Male": 1 2 2 2 1 1 2 1 1 2 ...
 $ PPAGE       : int  18 56 50 33 65 70 55 22 25 55 ...
 $ ppagecat    : Factor w/ 7 levels "18-24","25-34",..: 1 5 4 2 6 6 5 1 2 5 ...
 $ PPETHM      : Factor w/ 5 levels "White, Non-Hispanic",..: 1 1 1 2 3 1 3 3 5 1 ...
 $ PPINCIMP    : Ord.factor w/ 19 levels "Less than $5,000"<..: 19 16 10 1 6 1 16 14 17 8 ...
 $ PPEDUC      : Factor w/ 14 levels "No formal education",..: 9 14 9 9 10 6 12 10 10 9 ...
 $ PPEDUCAT    : Factor w/ 4 levels "Less than high school",..: 2 4 2 2 3 1 4 3 3 2 ...
 $ work        : Factor w/ 2 levels "unemployed","employed": 1 2 2 1 1 1 2 2 1 1 ...
 $ PPWORK      : Factor w/ 7 levels "Not working - disabled",..: 4 6 6 1 5 1 7 6 4 5 ...
 $ marital     : Factor w/ 2 levels "single","partnered": 1 2 1 1 2 1 1 1 1 1 ...
 $ PPMARIT     : Factor w/ 6 levels "Never married",..: 1 3 1 1 3 6 5 1 1 5 ...
 $ PPMSACAT    : Factor w/ 2 levels "Metro","Non-Metro": 1 1 1 1 1 1 1 1 1 1 ...
 $ ppreg9      : Factor w/ 9 levels "East-North Central",..: 3 4 7 3 3 1 3 1 6 1 ...
 $ PPSTATEN    : Factor w/ 51 levels "AK","AL","AR",..: 32 4 46 39 35 16 32 15 5 23 ...
 $ PPHOUSE     : Factor w/ 5 levels "A building with 2 or more apartments",..: 4 4 1 1 1 4 1 4 4 1 ...
 $ PPRENT      : Factor w/ 3 levels "Occupied without payment of cash rent",..: 2 2 3 3 3 3 2 2 2 3 ...
 $ PPNET       : Factor w/ 2 levels "No","Yes": 2 2 2 1 1 2 1 2 2 1 ...
 $ Q7_otherText: Factor w/ 15 levels "bike,walk","car",..: NA NA NA NA NA NA NA NA NA NA ...
 $ weight      : num  1.314 0.646 0.799 1.642 1.223 ...
 $ Q7_q        : Factor w/ 7 levels "Airplane","Bus",..: 2 2 2 2 2 2 2 2 2 2 ...
 $ Q7_r        : chr  "Yes" "Yes" "Yes" "Yes" ...

# survey design
options(digits = 4)
options(survey.lonely.psu = "adjust")
des7 <- svydesign(ids = ~1, weights = ~weight, data = q7_df[is.na(q7_df$weight)==F, ])

Gender, age, ethnicity, income, education, work

# weighted data frame
q7 <- data.frame(svytable(~Q7_q + Q7_r + PPGENDER + ppagecat + PPETHM + PPINCIMP +
                            PPEDUC + PPEDUCAT + work + PPWORK, des7, round = T))
# plot templates
title <- ggtitle("What types of public transportation do you regularly use?")

## main plot
p <- ggplot(q7, aes(Q7_q, weight = Freq)) + ptext
p + geom_bar() + title

## plot2: exclude 'Other' column
p2 <- ggplot(q7[!(q7$Q7_q)=='Other', ], aes(Q7_q, weight = Freq)) + ptext

# by gender
p2 + geom_bar() + aes(PPGENDER, fill = PPGENDER) + facet_wrap(~Q7_q) + ggtitle("By gender")

# age boxplot
svyboxplot(PPAGE~Q7_q, des7, main = "Age boxplot per response")

# by age group
p2 + geom_bar() + aes(ppagecat, fill = ppagecat) + facet_wrap(~Q7_q) + ggtitle("By age group")

p2 + geom_bar(position = "fill") + aes(ppagecat, fill = Q7_q)

# by ethnic group
p + geom_bar() + aes(PPETHM, fill = PPETHM) + facet_wrap(~Q7_q) + ggtitle("By ethnic group") +
  ptext2

p + geom_bar(position = "fill") + aes(PPETHM, fill = Q7_q)

p + geom_bar() + aes(fill = Q7_q) + facet_wrap(~PPETHM) + ggtitle("By ethnic group")

p + geom_bar(position = "fill") + aes(fill = PPETHM)

# by income
p + geom_bar() + aes(PPINCIMP, fill = PPINCIMP) + facet_wrap(~Q7_q) + ggtitle("By income") + 
  ptext2

p + geom_bar(position = "fill") + aes(PPINCIMP, fill = Q7_q)

# by education
p + geom_bar() + aes(PPEDUCAT, fill = PPEDUCAT) + facet_wrap(~Q7_q) + ggtitle("By education")

p + geom_bar(position = "fill") + aes(PPEDUCAT, fill = Q7_q)

p + geom_bar(position = "fill") + aes(PPEDUC, fill = Q7_q)

p + geom_bar() + aes(fill = Q7_q) + facet_wrap(~PPEDUCAT) + ggtitle("By education")

p + geom_bar(position = "fill") + aes(fill = PPEDUCAT)

# by work status
p + geom_bar() + aes(work, fill = work) + facet_wrap(~Q7_q) + ggtitle("By employment status")

p + geom_bar(position = "fill") + aes(PPWORK, fill = Q7_q) + ggtitle("By employment status")

Marital status, metro status, region, state of residency, house type, housing status, internet availability

# update weighted data frame
q7.2 <- data.frame(svytable(~Q7_q + Q7_r + marital + PPMARIT + PPMSACAT + ppreg9 +
                              PPSTATEN + PPHOUSE + PPRENT + PPNET, des7, round = T))

# restate plots
p3 <- ggplot(q7.2, aes(Q7_q, weight = Freq)) + ptext
p4 <- ggplot(q7.2[!(q7.2$Q7_q)=='Other', ], aes(Q7_q, weight = Freq)) + ptext

# by marital status
p3 + geom_bar() + aes(marital, fill = marital) + facet_wrap(~Q7_q) + ggtitle("By marital status")

p3 + geom_bar(position = "fill") + aes(marital, fill = Q7_q)

p3 + geom_bar() + aes(PPMARIT, fill = PPMARIT) + facet_wrap(~Q7_q)

p3 + geom_bar() + aes(PPMARIT, fill = Q7_q) + ggtitle("By marital status")

# by metro status
p3 + geom_bar() + aes(PPMSACAT, fill = PPMSACAT) + facet_wrap(~Q7_q) + ggtitle("By metro status")

p3 + geom_bar(position = "stack") + aes(fill = PPMSACAT)

p3 + geom_bar() + aes(fill = PPMSACAT) + facet_wrap(~PPMSACAT)

# by region
p3 + geom_bar(position = 'fill') + aes(fill = ppreg9) + ggtitle("Responses by US region")

p3 + geom_bar(position = 'fill') + aes(ppreg9, fill = Q7_q) + ggtitle("US regions by response")

# by state
p3 + geom_bar() + aes(PPSTATEN, fill = Q7_q) + coord_flip() + ggtitle("By state")

p3 + geom_bar(position = 'fill') + aes(PPSTATEN, fill = Q7_q) + coord_flip() + ggtitle("By state")

# by house type
p3 + geom_bar() + aes(fill = PPHOUSE) + ggtitle("By house type")

p3 + geom_bar(position = 'fill') + aes(fill = PPHOUSE)

p3 + geom_bar() + aes(PPHOUSE, fill = PPHOUSE) + facet_wrap(~Q7_q) + ptext2

# by housing status

# by internet availability
p3 + geom_bar(position = "fill") + aes(fill = PPNET) + ggtitle("Internet status")

p3 + geom_bar(position = 'fill') + aes(PPNET, fill = Q7_q) + ggtitle("Internet status")

Q8. For what types of activities do you regularly use public transportation?

# subset question data, rename columns, gather into single column
q8_df <- df %>%
  select(CaseID, PPGENDER, PPAGE, ppagecat, PPETHM, PPINCIMP, PPEDUC, PPEDUCAT,
         work, PPWORK, marital, PPMARIT, PPMSACAT, ppreg9, PPSTATEN, PPHOUSE, PPRENT, PPNET,
         Q8_1:Q8_6, weight) %>%
  rename(Work = Q8_1, School = Q8_2, Shopping = Q8_3, Visiting_people = Q8_4,
         Recreation = Q8_5, Other = Q8_6) %>%
  gather(Q8_q, Q8_r, Work:Other, na.rm = T) %>%
  mutate(Q8_q = as.factor(Q8_q))

# select only Yes responses
q8_df <- q8_df[q8_df$Q8_r == 'Yes', ]
str(q8_df)

'data.frame':   383 obs. of  21 variables:
 $ CaseID  : int  5 15 32 53 69 149 157 162 168 191 ...
 $ PPGENDER: Factor w/ 2 levels "Female","Male": 2 2 2 2 1 1 2 1 2 2 ...
 $ PPAGE   : int  56 50 27 55 22 26 48 25 31 25 ...
 $ ppagecat: Factor w/ 7 levels "18-24","25-34",..: 5 4 2 5 1 2 4 2 2 2 ...
 $ PPETHM  : Factor w/ 5 levels "White, Non-Hispanic",..: 1 1 5 3 3 5 2 1 4 1 ...
 $ PPINCIMP: Ord.factor w/ 19 levels "Less than $5,000"<..: 16 10 11 16 14 8 7 17 14 8 ...
 $ PPEDUC  : Factor w/ 14 levels "No formal education",..: 14 9 13 12 10 12 12 13 14 9 ...
 $ PPEDUCAT: Factor w/ 4 levels "Less than high school",..: 4 2 4 4 3 4 4 4 4 2 ...
 $ work    : Factor w/ 2 levels "unemployed","employed": 2 2 2 2 2 2 2 2 2 2 ...
 $ PPWORK  : Factor w/ 7 levels "Not working - disabled",..: 6 6 6 7 6 6 6 6 6 6 ...
 $ marital : Factor w/ 2 levels "single","partnered": 2 1 2 1 1 1 1 2 1 1 ...
 $ PPMARIT : Factor w/ 6 levels "Never married",..: 3 1 2 5 1 1 1 2 1 1 ...
 $ PPMSACAT: Factor w/ 2 levels "Metro","Non-Metro": 1 1 1 1 1 1 1 1 1 1 ...
 $ ppreg9  : Factor w/ 9 levels "East-North Central",..: 4 7 3 3 1 1 3 3 6 6 ...
 $ PPSTATEN: Factor w/ 51 levels "AK","AL","AR",..: 4 46 39 32 15 23 35 35 5 48 ...
 $ PPHOUSE : Factor w/ 5 levels "A building with 2 or more apartments",..: 4 1 1 1 4 4 1 1 1 4 ...
 $ PPRENT  : Factor w/ 3 levels "Occupied without payment of cash rent",..: 2 3 3 2 2 1 3 3 3 3 ...
 $ PPNET   : Factor w/ 2 levels "No","Yes": 2 2 2 1 2 1 2 2 2 2 ...
 $ weight  : num  0.646 0.799 0.384 1.264 1.813 ...
 $ Q8_q    : Factor w/ 6 levels "Other","Recreation",..: 6 6 6 6 6 6 6 6 6 6 ...
 $ Q8_r    : chr  "Yes" "Yes" "Yes" "Yes" ...

# survey design
options(digits = 4)
options(survey.lonely.psu = "adjust")
des8 <- svydesign(ids = ~1, weights = ~weight, data = q8_df[is.na(q8_df$weight)==F, ])

Gender, age, ethnicity, income, education

# weighted data frame
q8 <- data.frame(svytable(~Q8_q + Q8_r + PPGENDER + ppagecat + PPETHM + PPINCIMP + 
                            PPEDUC + PPEDUCAT + work + PPWORK, des8, round = T))
# plot templates
title <- ggtitle("For what types of activities do you regularly use public transportation?")

## main plot
p <- ggplot(q8, aes(Q8_q, weight = Freq)) + ptext
p + geom_bar() + title

## plot2: exclude 'Other' column
p2 <- ggplot(q8[!(q8$Q8_q)=='Other', ], aes(Q8_q, weight = Freq)) + ptext

# by gender
p2 + geom_bar() + aes(PPGENDER, fill = PPGENDER) + facet_wrap(~Q8_q) + ggtitle("By gender")

# age boxplot
svyboxplot(PPAGE~Q8_q, des8, main = "Age boxplot per response")

# by age group
p2 + geom_bar() + aes(ppagecat, fill = ppagecat) + facet_wrap(~Q8_q) + ggtitle("By age group")

p2 + geom_bar(position = "fill") + aes(ppagecat, fill = Q8_q)

# by ethnic group
p + geom_bar() + aes(PPETHM, fill = PPETHM) + facet_wrap(~Q8_q) + ggtitle("By ethnic group") +
  ptext2

p + geom_bar(position = "fill") + aes(PPETHM, fill = Q8_q)

p + geom_bar() + aes(fill = Q8_q) + facet_wrap(~PPETHM) + ggtitle("By ethnic group")

p + geom_bar(position = "fill") + aes(fill = PPETHM)

# by income
p + geom_bar() + aes(PPINCIMP, fill = PPINCIMP) + facet_wrap(~Q8_q) + ggtitle("By income") + 
  ptext2

p + geom_bar(position = "fill") + aes(PPINCIMP, fill = Q8_q)

# by education
p + geom_bar() + aes(PPEDUCAT, fill = PPEDUCAT) + facet_wrap(~Q8_q) + ggtitle("By education")

p + geom_bar(position = "fill") + aes(PPEDUCAT, fill = Q8_q)

p + geom_bar(position = "fill") + aes(PPEDUC, fill = Q8_q)

p + geom_bar() + aes(fill = Q8_q) + facet_wrap(~PPEDUCAT) + ggtitle("By education")

p + geom_bar(position = "fill") + aes(fill = PPEDUCAT)

# by work status
p + geom_bar() + aes(work, fill = work) + facet_wrap(~Q8_q) + ggtitle("By employment status")

p + geom_bar(position = "fill") + aes(PPWORK, fill = Q8_q) + ggtitle("By employment status")

Marital status, metro status, region, state of residency, house type, housing status, internet availability

# weighted data frame
q8.2 <- data.frame(svytable(~Q8_q + Q8_r + work + marital + PPMARIT + PPMSACAT + ppreg9 +
                              PPSTATEN + PPHOUSE + PPRENT + PPNET, des8, round = T))
# restate plots
p3 <- ggplot(q8.2, aes(Q8_q, weight = Freq)) + ptext
p4 <- ggplot(q8.2[!(q8.2$Q8_q)=='Other', ], aes(Q8_q, weight = Freq)) + ptext

# by marital status
p3 + geom_bar() + aes(marital, fill = marital) + facet_wrap(~Q8_q) + ggtitle("By marital status")

p3 + geom_bar(position = "fill") + aes(marital, fill = Q8_q)

p3 + geom_bar() + aes(PPMARIT, fill = PPMARIT) + facet_wrap(~Q8_q)

p3 + geom_bar() + aes(PPMARIT, fill = Q8_q) + ggtitle("By marital status")

# by metro status
p3 + geom_bar() + aes(PPMSACAT, fill = PPMSACAT) + facet_wrap(~Q8_q) + ggtitle("By metro status")

p3 + geom_bar(position = "stack") + aes(fill = PPMSACAT)

p3 + geom_bar() + aes(fill = PPMSACAT) + facet_wrap(~PPMSACAT)

# by region
p3 + geom_bar(position = 'fill') + aes(fill = ppreg9) + ggtitle("Responses by US region")

p3 + geom_bar(position = 'fill') + aes(ppreg9, fill = Q8_q) + ggtitle("US regions by response")

# by state
p3 + geom_bar() + aes(PPSTATEN, fill = Q8_q) + coord_flip() + ggtitle("By state")

p3 + geom_bar(position = 'fill') + aes(PPSTATEN, fill = Q8_q) + coord_flip() + ggtitle("By state")

# by house type
p3 + geom_bar() + aes(fill = PPHOUSE) + ggtitle("By house type")

p3 + geom_bar(position = 'fill') + aes(fill = PPHOUSE)

p3 + geom_bar() + aes(PPHOUSE, fill = PPHOUSE) + facet_wrap(~Q8_q) + ptext2

# by housing status

# by internet availability
p3 + geom_bar(position = "fill") + aes(fill = PPNET) + ggtitle("Internet status")

p3 + geom_bar(position = 'fill') + aes(PPNET, fill = Q8_q) + ggtitle("Internet status")

Q9. Do other members of your household regularly use public transportation?

Ethnicity, income, metro status, region, state, house type, housing status

# weighted data frame
q9 <- data.frame(svytable(~Q9 + PPETHM + PPINCIMP + PPMSACAT + ppreg9 + 
                            PPSTATEN + PPHOUSE + PPRENT, des, round = T))
# plot templates
title <- ggtitle("Do other members of your household regularly use public transportation?")

## main plot
p <- ggplot(q9, aes(Q9, weight = Freq)) + ptext
p + geom_bar() + title

## plot2: exclude 'Don_t know' column
p2 <- ggplot(q9[!(q9$Q9)=='Don_t know', ], aes(Q9, weight = Freq)) + ptext

# by ethnic group
p2 + geom_bar() + aes(fill = PPETHM) + ggtitle("By ethnic group") + ptext2

p2 + geom_bar(position = "fill") + aes(PPETHM, fill = Q9)

p2 + geom_bar() + aes(fill = Q9) + facet_wrap(~PPETHM) + ggtitle("By ethnic group")

p2 + geom_bar(position = "fill") + aes(fill = PPETHM)

# by income
p2 + geom_bar() + aes(fill = PPINCIMP) + ggtitle("By income") + ptext2

p2 + geom_bar(position = "fill") + aes(PPINCIMP, fill = Q9)

# by metro status
p2 + geom_bar() + aes(fill = PPMSACAT) + ggtitle("By metro status")

p2 + geom_bar(position = "stack") + aes(PPMSACAT, fill = Q9)

# by region
p2 + geom_bar(position = 'fill') + aes(fill = ppreg9) + ggtitle("Responses by US region")

p2 + geom_bar(position = 'fill') + aes(ppreg9, fill = Q9) + ggtitle("US regions by response")

# by state
p2 + geom_bar() + aes(PPSTATEN, fill = Q9) + coord_flip() + ggtitle("By state")

p2 + geom_bar(position = 'fill') + aes(PPSTATEN, fill = Q9) + coord_flip() + ggtitle("By state")

# by house type
p2 + geom_bar() + aes(fill = PPHOUSE) + ggtitle("By housing")

p2 + geom_bar(position = 'fill') + aes(fill = PPHOUSE)

p2 + geom_bar() + aes(PPHOUSE, fill = Q9) + ptext2

# by housing status

Q10. What types of public transportation do other members of your household regularly use?

# subset question data, rename columns, gather into single column
q10_df <- df %>%
  select(CaseID, PPGENDER, PPAGE, ppagecat, PPETHM, PPINCIMP, PPEDUC, PPEDUCAT,
         work, PPWORK, marital, PPMARIT, PPMSACAT, ppreg9, PPSTATEN, PPHOUSE, PPRENT, PPNET,
       Q10_1:Q10_8, weight) %>%
  rename(Bus = Q10_1, Carpool = Q10_2, Subway = Q10_3, Train = Q10_4,
         Taxi = Q10_5, Airplane = Q10_6, Don_t_know = Q10_7, Other = Q10_8) %>%
  gather(Q10_q, Q10_r, Bus:Other, na.rm = T) %>%
  mutate(Q10_q = as.factor(Q10_q))

# select only Yes responses
q10_df <- q10_df[q10_df$Q10_r == 'Yes', ]
str(q10_df)

'data.frame':   308 obs. of  21 variables:
 $ CaseID  : int  26 44 61 69 72 127 137 145 162 177 ...
 $ PPGENDER: Factor w/ 2 levels "Female","Male": 1 1 1 1 1 1 1 2 1 1 ...
 $ PPAGE   : int  37 70 45 22 25 75 59 21 25 31 ...
 $ ppagecat: Factor w/ 7 levels "18-24","25-34",..: 3 6 4 1 2 7 5 1 2 2 ...
 $ PPETHM  : Factor w/ 5 levels "White, Non-Hispanic",..: 1 1 1 3 5 2 2 1 1 1 ...
 $ PPINCIMP: Ord.factor w/ 19 levels "Less than $5,000"<..: 13 1 16 14 17 14 1 13 17 18 ...
 $ PPEDUC  : Factor w/ 14 levels "No formal education",..: 11 6 9 10 10 4 1 10 13 12 ...
 $ PPEDUCAT: Factor w/ 4 levels "Less than high school",..: 3 1 2 3 3 1 1 3 4 4 ...
 $ work    : Factor w/ 2 levels "unemployed","employed": 2 1 2 2 1 1 1 2 2 2 ...
 $ PPWORK  : Factor w/ 7 levels "Not working - disabled",..: 6 1 6 6 4 5 2 6 6 6 ...
 $ marital : Factor w/ 2 levels "single","partnered": 2 1 2 1 1 1 1 1 2 2 ...
 $ PPMARIT : Factor w/ 6 levels "Never married",..: 3 6 3 1 1 4 1 1 2 3 ...
 $ PPMSACAT: Factor w/ 2 levels "Metro","Non-Metro": 2 1 1 1 1 1 1 1 1 1 ...
 $ ppreg9  : Factor w/ 9 levels "East-North Central",..: 1 1 3 1 6 3 9 4 3 6 ...
 $ PPSTATEN: Factor w/ 51 levels "AK","AL","AR",..: 16 16 35 15 5 35 3 45 35 48 ...
 $ PPHOUSE : Factor w/ 5 levels "A building with 2 or more apartments",..: 4 4 4 4 4 1 4 4 1 4 ...
 $ PPRENT  : Factor w/ 3 levels "Occupied without payment of cash rent",..: 2 3 2 2 2 3 3 2 3 2 ...
 $ PPNET   : Factor w/ 2 levels "No","Yes": 2 2 2 2 2 2 1 2 2 2 ...
 $ weight  : num  0.935 0.957 0.845 1.813 0.446 ...
 $ Q10_q   : Factor w/ 8 levels "Airplane","Bus",..: 2 2 2 2 2 2 2 2 2 2 ...
 $ Q10_r   : chr  "Yes" "Yes" "Yes" "Yes" ...

# survey design
options(digits = 4)
options(survey.lonely.psu = "adjust")
des10 <- svydesign(ids = ~1, weights = ~weight, data = q10_df[is.na(q10_df$weight)==F, ])

Gender, age, ethnicity, income, education, work

# weighted data frame
q10 <- data.frame(svytable(~Q10_q + Q10_r + PPGENDER + ppagecat + PPETHM + PPINCIMP +
                            PPEDUC + PPEDUCAT + work + PPWORK, des10, round = T))
# plot templates
title <- ggtitle("What types of public transportation do other members of your household regularly use?")

## main plot
p <- ggplot(q10, aes(Q10_q, weight = Freq)) + ptext
p + geom_bar() + title

## plot2: exclude 'Don_t know' column
p2 <- ggplot(q10[!(q10$Q10_q)=='Don_t_know', ], aes(Q10_q, weight = Freq)) + ptext

# by gender
p2 + geom_bar() + aes(PPGENDER, fill = PPGENDER) + facet_wrap(~Q10_q) + ggtitle("By gender")

# by age group
p2 + geom_bar() + aes(ppagecat, fill = ppagecat) + facet_wrap(~Q10_q) + ggtitle("By age group")

p2 + geom_bar(position = "fill") + aes(ppagecat, fill = Q10_q)

# by ethnic group
p + geom_bar() + aes(PPETHM, fill = PPETHM) + facet_wrap(~Q10_q) + ggtitle("By ethnic group") +
  ptext2

p + geom_bar(position = "fill") + aes(PPETHM, fill = Q10_q)

p + geom_bar() + aes(fill = Q10_q) + facet_wrap(~PPETHM) + ggtitle("By ethnic group")

p + geom_bar(position = "fill") + aes(fill = PPETHM)

# by income
p + geom_bar() + aes(PPINCIMP, fill = PPINCIMP) + facet_wrap(~Q10_q) + ggtitle("By income") + 
  ptext2

p + geom_bar(position = "fill") + aes(PPINCIMP, fill = Q10_q)

# by education
p + geom_bar() + aes(PPEDUCAT, fill = PPEDUCAT) + facet_wrap(~Q10_q) + ggtitle("By education")

p + geom_bar(position = "fill") + aes(PPEDUCAT, fill = Q10_q)

p + geom_bar(position = "fill") + aes(PPEDUC, fill = Q10_q)

p + geom_bar() + aes(fill = Q10_q) + facet_wrap(~PPEDUCAT) + ggtitle("By education")

p + geom_bar(position = "fill") + aes(fill = PPEDUCAT)

# by work status
p + geom_bar() + aes(work, fill = work) + facet_wrap(~Q10_q) + ggtitle("By employment status")

p + geom_bar(position = "fill") + aes(PPWORK, fill = Q10_q) + ggtitle("By employment status")

Marital status, metro status, region, state of residency, house type, housing status, internet availability

# update weighted data frame
q10.2 <- data.frame(svytable(~Q10_q + Q10_r + marital + PPMARIT + PPMSACAT + ppreg9 +
                              PPSTATEN + PPHOUSE + PPRENT + PPNET, des10, round = T))
# restate plots
p3 <- ggplot(q10.2, aes(Q10_q, weight = Freq)) + ptext
p4 <- ggplot(q10.2[!(q10.2$Q10_q)=='Don_t_know', ], aes(Q10_q, weight = Freq)) + ptext

# by marital status
p3 + geom_bar() + aes(marital, fill = marital) + facet_wrap(~Q10_q) + ggtitle("By marital status")

p3 + geom_bar(position = 'fill') + aes(marital, fill = Q10_q)

p3 + geom_bar() + aes(PPMARIT, fill = PPMARIT) + facet_wrap(~Q10_q)

p3 + geom_bar() + aes(PPMARIT, fill = Q10_q) + ggtitle("By marital status")

# by metro status
p3 + geom_bar() + aes(PPMSACAT, fill = PPMSACAT) + facet_wrap(~Q10_q) + ggtitle("By metro status")

p3 + geom_bar(position = 'stack') + aes(fill = PPMSACAT)

p3 + geom_bar() + aes(fill = PPMSACAT) + facet_wrap(~PPMSACAT)

# by region
p3 + geom_bar(position = 'fill') + aes(fill = ppreg9) + ggtitle("Responses by US region")

p3 + geom_bar(position = 'fill') + aes(ppreg9, fill = Q10_q) + ggtitle("US regions by response")

# by state
p3 + geom_bar() + aes(PPSTATEN, fill = Q10_q) + coord_flip() + ggtitle("By state")

p3 + geom_bar(position = 'fill') + aes(PPSTATEN, fill = Q10_q) + coord_flip() + ggtitle("By state")

# by house type
p3 + geom_bar() + aes(fill = PPHOUSE) + ggtitle("By housing")

p3 + geom_bar(position = 'fill') + aes(fill = PPHOUSE)

p3 + geom_bar() + aes(PPHOUSE, fill = PPHOUSE) + facet_wrap(~Q10_q) + ptext2

# by housing status

# by internet availability
p3 + geom_bar(position = 'fill') + aes(fill = PPNET) + ggtitle("Internet status")

p3 + geom_bar(position = 'fill') + aes(PPNET, fill = Q10_q) + ggtitle("Internet status")